Generating realistic building electrical load profiles through the Generative Adversarial Network (GAN)

Wang, Zhe; Hong, Tianzhen

doi:10.1016/j.enbuild.2020.110299

Cited by 82 publications

(19 citation statements)

References 84 publications

(126 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Electrical consumption of buildings depends on socioeconomic aspects, on day type and on the number of occupants, but also by its surroundings and local climate conditions able to affect for example the daylight [13][14][15]. Regarding the PV production on rooftops [16], the solar energy potential depends on the suitable roof area available, on the roof slope, and on the roof orientation (south-faced tilted roofs have a higher productivity).…”

Section: Methodsmentioning

confidence: 99%

Towards Energy Self-consumption and Self-sufficiency in Urban Energy Communities

Todeschi¹,

Marocco²,

Mutani³

et al. 2021

IJHT

View full text Add to dashboard Cite

In Europe, 70% of citizens live in urban areas and consume around 75% of the primary energy supply. In order to reduce the impact of energy consumption and improve the competitiveness of local energy systems, Energy Communities may help to address the challenges of urban sustainability and energy security through local energy production and self-consumption. Solar, biomass and wind are the main sources of renewable energy that are generally used in cities. However, not all the sources available in urban environment are usable, due to the limited availability, or other technical or non-technical limits and constraints. In order to promote renewable energy technologies in buildings it is necessary to consider architectural, cultural, energy, technical and economic feasibility. This work defines a methodology for the optimal design of grid connected PV-battery systems in urban environments. The model was applied to two districts located in the city of Turin with the aim of evaluating the technical feasibility of combining multiple residential users at city level. The purpose of this work is to promote self-consumption and self-sufficiency from the network, using the integration of solar energy with PV-battery systems, and to reduce electrical losses in favor of both the single user and the distribution system. Results show that different values of self-sufficiency and self-consumption can be reached depending on the shape and dimension of each building. It was shown that it is possible to satisfy the current requirements to become an Energy Community in an urban environment with good levels of self-sufficiency.

show abstract

Section: Methodsmentioning

confidence: 99%

Towards Energy Self-consumption and Self-sufficiency in Urban Energy Communities

Todeschi¹,

Marocco²,

Mutani³

et al. 2021

IJHT

View full text Add to dashboard Cite

show abstract

“…The procedure stops when the discriminator is not able to discriminate whether the objects are generated or real. In a recent work [42], GAN was applied to one year of hourly whole building electrical meter data from 156 office buildings so that the individual variations of each building were eliminated. The generated load profiles were close to the real ones suggesting that GAN can be further used to anonymize data, generate load profiles and verify other generation models.…”

Section: Learning-based Methodsmentioning

confidence: 99%

An Approach to Data Acquisition for Urban Building Energy Modeling Using a Gaussian Mixture Model and Expectation-Maximization Algorithm

2021

View full text Add to dashboard Cite

In recent years, a building’s energy performance is becoming uncertain because of factors such as climate change, the Covid-19 pandemic, stochastic occupant behavior and inefficient building control systems. Sufficient measurement data is essential to predict and manage a building’s performance levels. Assessing energy performance of buildings at an urban scale requires even larger data samples in order to perform an accurate analysis at an aggregated level. However, data are not only expensive, but it can also be a real challenge for communities to acquire large amounts of real energy data. This is despite the fact that inadequate knowledge of a full population will lead to biased learning and the failure to establish a data pipeline. Thus, this paper proposes a Gaussian mixture model (GMM) with an Expectation-Maximization (EM) algorithm that will produce synthetic building energy data. This method is tested on real datasets. The results show that the parameter estimates from the model are stable and close to the true values. The bivariate model gives better performance in classification accuracy. Synthetic data points generated by the models show a consistent representation of the real data. The approach developed here can be useful for building simulations and optimizations with spatio-temporal mapping.

show abstract

“…4) Statistical parameters: [19] quantifies the load shape of a building's daily electricity consumption using five essential parameters:…”

Section: B Performance Metricsmentioning

confidence: 99%

Variational Autoencoder Generative Adversarial Network for Synthetic Data Generation in Smart Home

Razghandi¹,

Zhang²,

Erol‐Kantarci³

et al. 2022

Preprint

View full text Add to dashboard Cite

Data is the fuel of data science and machine learning techniques for smart grid applications, similar to many other fields. However, the availability of data can be an issue due to privacy concerns, data size, data quality, and so on. To this end, in this paper, we propose a Variational AutoEncoder Generative Adversarial Network (VAE-GAN) as a smart grid data generative model which is capable of learning various types of data distributions and generating plausible samples from the same distribution without performing any prior analysis on the data before the training phase. We compared the Kullback-Leibler (KL) divergence, maximum mean discrepancy (MMD), and Wasserstein distance between the synthetic data (electrical load and PV production) distribution generated by the proposed model, vanilla GAN network, and the real data distribution, to evaluate the performance of our model. Furthermore, we used five key statistical parameters to describe the smart grid data distribution and compared them between synthetic data generated by both models and real data. Experiments indicate that the proposed synthetic data generative model outperforms the vanilla GAN network. The distribution of VAE-GAN synthetic data is the most comparable to that of real data.

show abstract

Generating realistic building electrical load profiles through the Generative Adversarial Network (GAN)

Cited by 82 publications

References 84 publications

Towards Energy Self-consumption and Self-sufficiency in Urban Energy Communities

Towards Energy Self-consumption and Self-sufficiency in Urban Energy Communities

An Approach to Data Acquisition for Urban Building Energy Modeling Using a Gaussian Mixture Model and Expectation-Maximization Algorithm

Variational Autoencoder Generative Adversarial Network for Synthetic Data Generation in Smart Home

Contact Info

Product

Resources

About